Data-aided chromatic dispersion estimation

Abstract

The present invention relates to performing chromatic dispersion estimation in a receiver of an optical communication system. Here, the signal received by the receiver includes frames, each comprising a training portion and a data portion. The training portion comprises a plurality of identical pattern sequences. Different settings are applied to an equalizer to generate a plurality of equalized signals from at least one of the received frames. Then, at least one correlation value is calculated between a first pattern sequence and a second pattern sequence of the equalized signals and a final correlation value is derived from the respective correlation values. The setting of the equalizer corresponding to the equalized signal providing the highest final correlation value is selected to provide the chromatic dispersion estimation.

The invention claimed is: 
     
       1. A method of selecting a setting of an equalizer in a receiver of an optical communication system for compensating for chromatic dispersion in a received signal, comprising the steps of:
 receiving a signal divided into frames, wherein each frame comprises a training portion and a data portion, and wherein the training portion comprises at least two identical pattern sequences, 
 equalizing the received signal such as to generate a plurality of equalized signals, each corresponding to a different setting of an equalizer, wherein each of the equalized signals comprises at least one frame, 
 determining a plurality of correlation values, wherein at least one correlation value is calculated for each frame comprising the equalized signal, and wherein each correlation value is calculated between a first pattern sequence and a second pattern sequence of the corresponding frame, 
 determining a final correlation value for each of the equalized signals by averaging the plurality of correlation values determined for each of the equalized signals, and 
 selecting the setting of the equalizer corresponding to the equalized signal providing a highest final correlation value. 
 
     
     
       2. The method according to  claim 1 , wherein determining at least one correlation value for each of the equalized signals comprises:
 generating a delayed version of each of the equalized signals by delaying each equalized signal constituting a non-delayed version by a time-offset, 
 determining at least one correlation value for each frame of the delayed and non-delayed versions of each of the equalized signals by calculating the correlation between a first pattern sequence of the delayed version of the equalized signal and a second pattern sequence of the version of the non-delayed equalized signal. 
 
     
     
       3. The method according to  claim 2 , wherein determining at least one correlation value or each of the equalized signals comprises multiplying samples of the non-delayed version of the equalized signal with samples of the delayed version of the equalized signal and calculating a moving average of the multiplied samples. 
     
     
       4. The method according to  claim 3 , wherein calculating the moving average provides a sequence of power values and wherein a correlation value is determined as a peak power value of a portion of the sequence of power values. 
     
     
       5. The method according to  claim 2 , wherein the time-offset used to delay the equalized signal corresponds to a multiple of the length of a pattern sequence in the training portion of a frame. 
     
     
       6. The method according to  claim 1 , wherein the pattern sequences comprised in the training portion of a frame represent training sequences used for channel estimation and/or frame detection and/or frequency offset estimation. 
     
     
       7. The method according to  claim 1 , wherein the equalizer is a frequency domain equalizer. 
     
     
       8. The method according to  claim 1 , wherein frames of the received signal comprise symbols having different amplitudes and phasesor wherein the received signal represents a polarization-division multiplexed (PDM) signal. 
     
     
       9. The method according to  claims 1 , further comprising the step of calculating the correlation values by a frame detection unit. 
     
     
       10. The method according to  claim 1 , further comprising the step of calculating the correlation values by a frequency offset estimation unit. 
     
     
       11. The method according to  claim 1 , wherein for at least one of the frames of the received signal the training portion comprises non-contiguous portions separated by parts of the data portion. 
     
     
       12. A receiver for an optical communication system, comprising:
 an input port adapted to receive a signal divided into frames, wherein each frame comprises a training portion and a data portion, and wherein the training portion comprises at least two identical pattern sequences, 
 an equalizer adapted to equalize the received signal to generate a plurality of equalized signals, each corresponding to a different setting of the equalizer, 
 a correlation unit adapted to determine correlation values for each of the equalized signals by calculating each correlation value between a first pattern sequence and a second pattern sequence of each frame included in the equalized signal, 
 an averaging unit adapted to provide a final correlation value by averaging the correlation values calculated for each of the equalized signals, and 
 a selection unit adapted to select the setting of the equalizer corresponding to the equalized signal providing a highest final correlation value. 
 
     
     
       13. The receiver according to  claim 12 , further comprising a frame detection unit configured to receive the equalized signal provided by the equalizer, wherein the frame detection unit is preferably adapted to detect the training portion, calculate the correlation between the training portion and a delayed version of the training portion to determine power values, and wherein the frame detection unit is further adapted to determine the correlation value corresponding to the peak power value of the determined power values and to output the determined correlation value to the selection unit. 
     
     
       14. The receiver according to  claim 12 , further comprising computing circuits which are shared between the correlation unit and one of the frame detection unit or a frequency offset estimation unit. 
     
     
       15. The receiver ( 100 ) according to  claim 12 , wherein the equalizer is adapted to compensate chromatic dispersion (CD) in a received polarization-division multiplexed (PDM) signal.